4 ms·

I thought it would be mayhem but traffic overall didn't change at all. It's exactly the same.

Areading314 7y ago
I had a car for several years in the city and never drove down market a single time. Always faster to take the one-ways mission or howard.
wahern 7y ago
That's because most cars were taken off Market St in 2015: https://www.sfmta.com/project-updates/turn-restrictions-market-street-going-effect-aug-11-2015 https://www.sfmta.com/project-updates/turn-restrictions-mark... Since then most cars on Market St north of Van Ness seem to have been Uber and Lyft. Now they're finally banned. If buses are actually running faster--not just a consequence of less traffic during the winter months--it goes to show how much Uber and Lyft have contributed to traffic in the city. They drive slow, erratically, and stop in the middle of the road on main arteries to pick people up and drop them off.
- Areading314 7y ago
  It still doesn't make sense that they let the "yellow" taxis continue driving on market. They are on average much worse drivers
  - new_realist 7y ago
    There are fewer of them and they are more efficient, running at higher passenger utilization.
    - thedance 7y ago
      This can’t be true. Taxis often run with zero passengers, whereas a private car never does.
      - wahern 7y ago
        I could only find two studies on capacity utilization: * Judd Cramer, Alan B. Krueger, "Disruptive Change in the Taxi Business: The Case of Uber", 2016, https://www.nber.org/papers/w22083 https://www.nber.org/papers/w22083 and * Y.M. Nie, "How can the taxi industry survive the tide of ridesourcing? Evidence from Shenzhen, China", 2017 https://www.sciencedirect.com/science/article/pii/S0968090X17301018 https://www.sciencedirect.com/science/article/pii/S0968090X1... Both seem to confirm that ride-hailing has better utilization rates, though Nie qualifies that--e.g. sometimes too many private cars might turn out. However, at least in San Francisco all the licensed taxi cabs also use ride-hailing apps, so their utilization should be at least as good as unlicensed cars. The real issue with ride-hailing is induced demand. In places like San Francisco public transit took a huge dip, and ride-hailing simply isn't anywhere as efficient as mass transit. Studies have shown that ride-hailing has contributed to a substantial increase in traffic and traffic delay: https://advances.sciencemag.org/content/5/5/eaau2670 https://advances.sciencemag.org/content/5/5/eaau2670 Ride-hailing can even be worse than personal cars because a personal car has a 100% utilization rate at minimum, whereas the average utilization rate for ride-hailing, even accounting for multiple riders (e.g. >100% for any particular trip) is <100%. Ride-hailing only pencils out for the last mile--to and from transit hubs in a local vicinity. But too many people take it to cross town. It's hard to blame ride-hailing companies, though. Cities need to be build more subways. Perhaps companies like Uber, if they were allowed to build subways as cheaply as The Boring Company can do (i.e. small diameter tunnel, though still larger than London and Budapest; no red tape; no extravagant stations) and to capture the revenue, would invest the money. They'd finally be able to ditch drivers, too, because driverless cars will probably only ever work, if they work at all, in moderately dense residential areas.
        eru 7y ago
        Personal car usage doesn't have a 100% utilization: lots of time is spent cruising for parking and coming back from some out of the way parking. You can't really count that as being productively utilized. It's the opposite.
      - baddox 7y ago
        It can be true if enough taxi time is spent with 2 or more passengers.
  - carapace 7y ago
    > They are on average much worse drivers How do you know?
- eru 7y ago
  It's a bit weird that some people (not you) single out Uber and Lyft when talking about congestion. A congestion charge would solve the problem in a principled way, with no need for exceptions for special vehicles. Buses and even fire engines and ambulances could just pay the congestion charge, too.
  - _jal 7y ago
    If you have a nearly full cup, and someone dumps a bunch more water at it, who do you blame for the mess?
    - scarejunba 7y ago
      The logical fallacy comes from the implicit nature of it being your cup rather than a shared cup. This is more like "if the park is crowded and a big family decided to celebrate their birthday there crowding it even more, who do you blame for the crowd?" Or "If you've got four bicycle stands and four of you bike to work but then a new guy joins and he bikes to work too". Of course you can still apportion blame/responsibility based on idle time, personal v commercial use, but who came first is less convincing when equal rights apply.
    - eru 7y ago
      Perhaps. Though I don't see how they are adding congestion? If people would take their own cars instead, that would congest the streets just as much, plus use up parking space.

I had a car for several years in the city and never drove down market a single time. Always faster to take the one-ways mission or howard.

wahern 7y ago

That's because most cars were taken off Market St in 2015: https://www.sfmta.com/project-updates/turn-restrictions-market-street-going-effect-aug-11-2015 https://www.sfmta.com/project-updates/turn-restrictions-mark... Since then most cars on Market St north of Van Ness seem to have been Uber and Lyft. Now they're finally banned. If buses are actually running faster--not just a consequence of less traffic during the winter months--it goes to show how much Uber and Lyft have contributed to traffic in the city. They drive slow, erratically, and stop in the middle of the road on main arteries to pick people up and drop them off.

Areading314 7y ago
It still doesn't make sense that they let the "yellow" taxis continue driving on market. They are on average much worse drivers
- new_realist 7y ago
  There are fewer of them and they are more efficient, running at higher passenger utilization.
  - thedance 7y ago
    This can’t be true. Taxis often run with zero passengers, whereas a private car never does.
    - wahern 7y ago
      I could only find two studies on capacity utilization: * Judd Cramer, Alan B. Krueger, "Disruptive Change in the Taxi Business: The Case of Uber", 2016, https://www.nber.org/papers/w22083 https://www.nber.org/papers/w22083 and * Y.M. Nie, "How can the taxi industry survive the tide of ridesourcing? Evidence from Shenzhen, China", 2017 https://www.sciencedirect.com/science/article/pii/S0968090X17301018 https://www.sciencedirect.com/science/article/pii/S0968090X1... Both seem to confirm that ride-hailing has better utilization rates, though Nie qualifies that--e.g. sometimes too many private cars might turn out. However, at least in San Francisco all the licensed taxi cabs also use ride-hailing apps, so their utilization should be at least as good as unlicensed cars. The real issue with ride-hailing is induced demand. In places like San Francisco public transit took a huge dip, and ride-hailing simply isn't anywhere as efficient as mass transit. Studies have shown that ride-hailing has contributed to a substantial increase in traffic and traffic delay: https://advances.sciencemag.org/content/5/5/eaau2670 https://advances.sciencemag.org/content/5/5/eaau2670 Ride-hailing can even be worse than personal cars because a personal car has a 100% utilization rate at minimum, whereas the average utilization rate for ride-hailing, even accounting for multiple riders (e.g. >100% for any particular trip) is <100%. Ride-hailing only pencils out for the last mile--to and from transit hubs in a local vicinity. But too many people take it to cross town. It's hard to blame ride-hailing companies, though. Cities need to be build more subways. Perhaps companies like Uber, if they were allowed to build subways as cheaply as The Boring Company can do (i.e. small diameter tunnel, though still larger than London and Budapest; no red tape; no extravagant stations) and to capture the revenue, would invest the money. They'd finally be able to ditch drivers, too, because driverless cars will probably only ever work, if they work at all, in moderately dense residential areas.
      - eru 7y ago
        Personal car usage doesn't have a 100% utilization: lots of time is spent cruising for parking and coming back from some out of the way parking. You can't really count that as being productively utilized. It's the opposite.
    - baddox 7y ago
      It can be true if enough taxi time is spent with 2 or more passengers.
- carapace 7y ago
  > They are on average much worse drivers How do you know?
eru 7y ago
It's a bit weird that some people (not you) single out Uber and Lyft when talking about congestion. A congestion charge would solve the problem in a principled way, with no need for exceptions for special vehicles. Buses and even fire engines and ambulances could just pay the congestion charge, too.
- _jal 7y ago
  If you have a nearly full cup, and someone dumps a bunch more water at it, who do you blame for the mess?
  - scarejunba 7y ago
    The logical fallacy comes from the implicit nature of it being your cup rather than a shared cup. This is more like "if the park is crowded and a big family decided to celebrate their birthday there crowding it even more, who do you blame for the crowd?" Or "If you've got four bicycle stands and four of you bike to work but then a new guy joins and he bikes to work too". Of course you can still apportion blame/responsibility based on idle time, personal v commercial use, but who came first is less convincing when equal rights apply.
  - eru 7y ago
    Perhaps. Though I don't see how they are adding congestion? If people would take their own cars instead, that would congest the streets just as much, plus use up parking space.